Tethys Svensson

> How do you feel about Rust 1.60 as the MSRV? I think it's probably the right choice in this situation, given that you are trying to stabilize. In planus,...

@sminez As far as I can tell, this might be the last remaining blocker before I can switch over to using penrose instead of xmonad (besides porting the functionality I...

:sweat_smile: Thanks you for the very detailed description. It makes a lot of sense that you would like to get tests improved before changing this part of the code. However...

But the `get_index_mut` function is not hidden behind the `MutableKeys` trait, but is directly on the `IndexMap`. Perhaps there should also be a `get_index_mut2` in the `MutableKeys` trait, and the...

> I suppose if the item already exists at any index, you want it moved to the new location? For my particular use-case I wish to keep the item at...

@tustvold Trying to fix **any** soundness issue is probably a bit of a high bar. :wink: There are still bugs in the parser like #7007.

I still think there are a lot of ways to bypass this. For instance this one: In `schema.fbs`: ``` table Foo { x: uint32; } ``` In `main.rs`: ```rust mod...

@tustvold It is, but the schema does not contain a string -- and the `get` function allows re-interpreting the data.

Personally I think a lot of these issues would be solved by doing late validation instead of using the verifier. That would also be faster, because you don't have to...

There is also the fact that `safe_slice` is inherently undefined behavior according to the Rust ABI, which e.g. requires `u32` to be 4-byte aligned. This is an example of this...